The Journal of Neuroscience, August 22, 2012 • 32(34):11773–11779 • 11773
Development/Plasticity/Repair
Effects of Temporary Functional Deafferentation on the
Brain, Sensation, and Behavior of Stroke Patients
Elisabeth Sens,1 Ulrike Teschner,1,2 Winfried Meissner,3 Christoph Preul,2 Ralph Huonker,2 Otto W. Witte,2
Wolfgang H. R. Miltner,1 and Thomas Weiss1
1Biological and Clinical Psychology, Friedrich Schiller University, Jena, D-07743, Germany, 2Department of Neurology, University Hospital, Jena, D-07747,
Germany, and 3Department of Anesthesiology and Intensive Care, University Hospital, Jena, D-07747, Germany
Following stroke, many patients suffer from chronic motor impairment and reduced somatosensation in the stroke-affected body parts.
Recent experimental studies suggest that temporary functional deafferentation (TFD) of parts of the stroke-affected upper limb or of the
less-affected contralateral limb might improve the sensorimotor capacity of the stroke-affected hand. The present study sought evidence
of cortical reorganization and related sensory and motor improvements following pharmacologically induced TFD of the stroke-affected
forearm.
Examination was performed during 2 d of Constraint-Induced Movement Therapy. Thirty-six human patients were deafferented on
the stroke-affected forearm by an anesthetic cream (containing lidocaine and prilocaine) on one of the 2 d, and a placebo cream was
applied on the other. The order of TFD and placebo treatment was counterbalanced across patients. Somatosensory and motor performance were assessed using a Grating orienting task and a Shape-sorter-drum task, and with somatosensory-evoked magnetic fields.
Evoked magnetic fields showed significant pre- to postevaluation magnitude increases in response to tactile stimulation of the thumb of
the stroke-affected hand during TFD but not following placebo treatment. We also observed a rapid extension of the distance between
cortical representations of the stroke-affected thumb and little finger following TFD but not following placebo treatment. Moreover,
somatosensory and motor performance of the stroke-affected hand was significantly enhanced during TFD but not during placebo
treatment. Thus, pharmacologically induced TFD of a stroke-affected forearm might improve the somatosensory and motor functions of
the stroke-affected upper limb, accompanied by cortical plasticity.
Introduction
The global burden of stroke is immense (Feigin et al., 2009).
Among consequences of stroke, motor impairment and reduced
somatosensation are common. In recent years the beneficial effects
of neurorehabilitative treatments and mechanisms of cortical plasticity have become more prominent in stroke rehabilitation
(Hummel et al., 2005; Cramer et al., 2011; Grefkes and Fink, 2011). A
new technique, able to modulate mechanisms of cortical reorganization by noninvasive means, is temporary functional deafferentation (TFD).
TFD has been demonstrated to improve somatosensory and
motor functions of the homonym contralateral or neighboring
ipsilateral body part and modulates their neural cortical representations. For example, TFD of the right hand by tourniquet-
Received Nov. 28, 2011; revised June 15, 2012; accepted June 23, 2012.
Author contributions: E.S., W.M., O.W.W., W.H.R.M., and T.W. designed research; E.S. and U.T. performed research; E.S., C.P., R.H., and T.W. analyzed data; E.S., U.T., W.M., C.P., R.H., O.W.W., W.H.R.M., and T.W. wrote the
paper.
This research was supported by grants from the Interdisciplinary Centre of Clinical Research (IZKF) Jena to
W.H.R.M. and T.W. Thanks to Stefan Clauss, Dr. Theresa Götz, and Tina Radtke for their contribution in MEG data
acquisition and analyses, and to Dr. Jeremy Thorne for language advice and manuscript editing.
The authors declare no competing financial interests.
Correspondence should be addressed to Prof. Thomas Weiss, Biological and Clinical Psychology, Friedrich Schiller
University, Am Steiger 3H1, Jena, D-07743, Germany. E-mail: thomas.weiss@uni-jena.de.
DOI:10.1523/JNEUROSCI.5912-11.2012
Copyright © 2012 the authors 0270-6474/12/3211773-07$15.00/0
induced anesthesia resulted in a rapid improvement of the grip
strength, tactile discrimination, and sensibility of the left hand
and increased the neural activity in the right primary motor cortex (Björkman et al., 2004b). Rossini et al. (1994) additionally
demonstrated an increase in neural dipole source activity and
deepening of the cortical representation of the middle finger during ischemic anesthesia of the neighboring fingers, explained by
unmasking of existing pathways in primary somatosensory cortex (SI). Furthermore, pharmacologically induced TFD of radial
and median nerves resulted in significant improvements in sensory discrimination around the ipsilateral lip. These improvements were accompanied by a shift of the cortical representations
of the ulnar skin area of the hand and lip into the deafferented
representations of parts of the hand, and a significant decrease of
intracortical inhibition within the muscle representation of the
extensor digiti minimi muscle (Weiss et al., 2004). Similarly,
Björkman et al. (2009) reported increased right hand sensitivity
and an expansion of its representation within SI in healthy subjects during TFD of the right forearm by a local anesthetic cream.
A few experimental studies have focused on TFD in chronic
stroke patients. For instance, Muellbacher et al. (2002) demonstrated improved hand motor function in stroke patients during
pharmacologically induced anesthesia. Furthermore, tourniquetinduced anesthesia of the less-affected hand was shown to improve
somatosensory sensibility (Voller et al., 2006) and the motor performance (Floel et al., 2004) of the affected hand. The improvement of
11774 • J. Neurosci., August 22, 2012 • 32(34):11773–11779
Sens et al. • Learning and Cortical Reorganization Induced by TFD
motor function was attributed to reduced
interhemispheric inhibition (Floel et al.,
2008). Weiss et al. (2011) demonstrated
beneficial effects of TFD on the somatosensory sensibility and motor capacity of the
stroke-affected hand using pharmacologically induced TFD of the more-affected
forearm during 1 d of Constraint-Induced
Movement Therapy (CIMT; Miltner et al.,
1999; Bauder et al., 2001; Langhorne et al., Figure 1. Time line of the experiment showing the time and order of examinations. Pre- and post-testing examinations were
2009). Together, these results suggest that identical.
TFD affects cortical reorganization and improves the somatosensory and motor ability
diameter enclosed in plastic housing that was fixed with adhesive plaster
of the stroke-affected upper limb. Previous reports of TFD in stroke
to the skin. The membrane pressed on the skin when expanded by an air
patients however have mainly focused on interhemispheric cortical
pressure pulse (25 psi). Stimuli were applied in random order with an
interstimulus interval ranging from 800 to 1600 ms (0.625–1.25 Hz). The
changes and plasticity of the primary motor cortex, and cortical reelectro-oculogram (EOG) was recorded from two vertically positioned
organization in SI on the ipsilesional (contralateral) hemisphere has
Ag/AgCl electrodes to identify recording epochs with eye movement
been less well studied.
artifacts. Trials with eye movement artifacts and with EOG amplitudes
In the present study we investigated changes of cortical actiexceeding 150 ␮V were excluded from further analysis. SEFs were samvation in SI following pharmacologically induced TFD of the
pled at a rate of 1000 Hz and recorded with a 0.1–330 Hz bandpass filter.
stroke-affected forearm during 1 d of CIMT, as well as its effects
An epoch lasted 500 ms, including a 100 ms prestimulus period. Off-line
on stroke patients’ somatosensory sensibility and motor capacity.
data analysis included Maxwell filtering (Maxfilter Version 2.0.21, using
Materials and Methods
Patients
Thirty-six chronic stroke patients (18 females, 18 males) who attended
CIMT at our treatment center at the Friedrich Schiller University participated in this study. Patients’ characteristics (age, lesion site, etc.) are
shown in Table 1. The procedure of the TFD experiment was extensively
described to participants who then provided written informed consent.
The procedure was conducted in accordance with the Helsinki Declaration on human experimentation and approved by the ethics committee
of the Friedrich Schiller University.
Experimental design
Similar to a previous examination (Weiss et al., 2011), subjects participated in the present experiment on 2 d of the standard CIMT (Bauder et
al., 2001). The experiment consisted of two separate treatments, the application of a cream that contained 20 g of a placebo agent, and a cream
containing 20 g of a eutectic anesthetic emulsion composed of 2.5%
lidocaine and 2.5% prilocaine (Emla, AstraZeneca). Both creams were
applied to the volar side of the more-affected forearm covering an area of
50 ⫻ 150 mm located ⬃10 mm above the wrist and sheathed with an
occlusive plaster. Each patient received both creams. The order of the two
treatments was counterbalanced across patients according to the order of
admission (even number—placebo first, odd number—TFD first).
Creams were applied in the morning before the start of the baseline
evaluation (t1) of magnetoencephalographic (MEG) measurements and
sensory and motor capacity assessment, and remained on the arm until
the end of the treatment evaluation on each day (t2). Subjects were told
that they would receive two types of a local anesthetic cream designed to
support the training. After the plaster was fixed to the forearm, baseline
MEG recordings of somatosensory evoked magnetic fields (SEFs) and the
tests outlined below were made (t1). The usual course of CIMT for 3.5 h
then followed. After the CIMT session, the same tests and SEF recordings
were performed again (t2; Fig. 1).
Recording of SEFs
SEFs were recorded using a whole-head 306 channel magnetoencephalograph (Elekta Neuromag). During SEF recordings, patients were lying
on a bed inside a magnetically shielded room with their head placed in the
mold of the dewar. Patients received ⬃300 non-noxious tactile stimuli
on the phalanges of the thumb (D1) and 300 similar stimuli on the little
finger (D5) of their stroke-affected hand using a pneumatically driven
stimulator (Somatosensory Stimulus Generator, 4-D NeuroImaging
Inc.). The stimulator consisted of a thin rubber membrane of 10 mm
the time domain extension), a baseline correction from ⫺90 to 0 ms, a
bandpass filter from 0.1 to 100 Hz, and selection of 50% of all channels,
i.e., channels over a single hemisphere. Before the SEF recordings, an
individual Cartesian head coordinate system was defined with a threedimensional digitizer to locate each patient’s head position in the MEG
dewar and to match the MEG data with magnetic resonance images
(MRIs). The x-axis connected the two periauricular points and the y-axis
was represented by the perpendicular projection from the nasion to the
midpoint of the x-axis. The z-axis was a line orthogonal to the crossing of
the x- and y-axis. Additionally, four position indicator coils were attached to the head and their locations digitized (Preissler et al., 2011).
An equivalent current dipole (ECD) model was used to estimate the
location and strength of the sources using a spherical head model, based
on the individual, T1 weighted MRI. The MRI recording was done on a
1.5 Tesla GYROSCAN system (Philips AG; 3D-Flash sequence, 256 sagittal MR images of 1 mm slice thickness, 256 ⫻ 256 matrix (1 ⫻ 1 mm
in-plane resolution) covering the entire head). For one patient a MRI
recording was not available, so the center of his sphere was set to the
Cartesian coordinates 0, 0, 40 (x, y, z).
The ECD of each stimulated finger was determined as the earliest
maximal activity between 35 and 80 ms poststimulus, excluding a 35 ms
delay of the pressure onset to the trigger. Dipole locations were accepted
if the goodness-of-fit was better than r ⫽ 0.85 and the dipole moment
exceeded 5 nA m. The Euclidean distance (ED) between the dipole locations of D1 and D5 was calculated for each patient and each MEG session
to quantify cortical reorganization of both fingers following TFD.
Changes in the amount of cortical activity were assessed by comparing
the dipole moments of D1 for each MEG examination. The dipole moment and the ED were transformed to a percentage of the pretest value to
standardize the data. MEG data were analyzed with DANA Software
Release 3 (Elekta).
Assessment of motor and sensory functions
Von-Frey hair testing. Von-Frey hair testing (VFHT) was used to characterize the efficiency of TFD on mechanical thresholds of the moreaffected forearm. Thresholds for touch were tested at a point marked for
VFHT assessment in the center of the occlusive bandage of the treated
area of subjects’ stroke-affected forearm. A von-Frey hair set (VF2 OptiHair 2; Marstock Nervtest) was used for sensory assessment. According
to Rolke et al. (2006), a method of limits with five ascending and descending series of VFHT was used to determine tactile detection thresholds.
VFHT was defined as the geometric mean of five suprathreshold and five
infrathreshold values. A normal distribution of VFHT was achieved by
logarithmic transformation of the parameter (log2 units) as the forces of
von-Frey hairs increase by a factor of 2 (Baumgärtner et al., 2002).
Sens et al. • Learning and Cortical Reorganization Induced by TFD
J. Neurosci., August 22, 2012 • 32(34):11773–11779 • 11775
Table 1. Characteristics of patients
Age
(years)
Number of Years after
Sex strokes stroke
NIHSS Lesion site
1
11
F
1
2.75
2
2
3
4
5
6
69
72
77
57
54
M
F
F
F
M
1
1
2
1
1
16.0
7.75
0.50
9.08
7.00
7
64
M
1
4.25
0
8
9
82
65
M
F
1
1
2.42
25.83
5
1
10
11
12
13
14
52
62
56
56
59
M 10
M 1
F
1
M 1
F
1
4.92
1.33
8.42
2.08
1.08
—
4
3
5
2
15
16
17
54
52
41
F
F
F
1
1
5
5.75
2.75
8.58
5
2
3
18
57
M
1
1.83
2
19
20
21
22
23
24
25
26
27
28
29
54
53
58
51
71
72
69
59
73
66
60
F
M
F
M
M
M
F
F
M
M
M
1
1
1
1
1
1
1
1
1
2
1
1.17
11.66
10.25
3.58
1.58
2.50
2.25
3.92
3.08
4.58
0.92
2
4
3
2
4
2
5
3
2
3
6
30
62
F
1
5.08
3
31
32
60
45
F
M
1
1
2.33
5.92
5
3
33
34
35
45
66
72
F
F
M
1
1
1
4.50
3.16
7.16
4
2
—
36
14
M
Mean ⫾ SD 58.06 ⫾ 14.53
1
Patient
4
—
3
7
6
1.83
—
5.22 ⫾ 4.94 3.0*
L–striolenticular infarction and infarction of prerolandic
artery territory
R–paramedian pontine infarction
SAE
L–infarction of rolandic artery territory
R–traumatic injury
L–MCI infarction after traumatic left internal carotid
artery dissection †
R–arteriosclerotic encephalopathy, right PICA infarction,
small lesion left internal capsule
R–striolenticular infarction
L–status post resection of an atypical intracerebral
hemorrhage
R–paramedian pontine infarction
L–thalamic lesion after abscess
L–striolenticular infarction
L–striolenticular infarction
L–multiple MCA-territory infarctions: A. gyri angularis, A.
frontobasalis lateralis, posterior striolenticular
territory
L–striolenticular infarction, initially with hemorrhage
R–striolenticular infarction
L–striolenticular DD, confluent microvascular lesion,
lateral white matter compartment
L–subcortical striolenticular territory lesion, initially with
(secondary) hemorrhage
R–striolenticular infarction
R–striolenticular plus suprainsular perisylvian infarction
L–striolenticular infarction
R–perisylvian MCI infarction
R–paramedian pontine infarction
R–paramedian pontine infarction
L–atypical MCA/ACA border zone infarction
L–subinsular residual hemorrhage
R–minor arterial branch infarction
L–paramedian pontine infarction
L– capsular lacunar lesion, R–thalamic lacunar lesion,
R–pontine lacunar lesion
R–mainly striolenticular infarction with cortical insular
involvement
R–paramedian pontine infarction
L–striolenticular infarction with cortical insular
involvement
R-total MCI infarction
R–striolenticular/anterior MCA territory infarction
R–striolenticular infarction with cortical insular
involvement
R–lacunar thalamic lesion
WMFT
MAL
Sensory
deficit
Mainly subcortical
2.9
0.30
0.92
Subcortical
Cortical
Cortical and subcortical
Cortical and subcortical
3.8
3.5
3.7
3.1
4.0
1.24
3.24
4.28
0.17
2.20
2.29
2.67
1.05
6.15
1.50
Multi infarct
3.7
2.40
1.88
Subcortical
3.8
Cortical and subcortical 2.4
1.33
1.10
0.73
1.08
Subcortical
Subcortical
Subcortical
Cortical and subcortical
2.9
3.6
3.7
2.1
4.0
1.53
3.07
2.11
0.41
4.10
1.00
1.43
1.43
1.25
0.79
Subcortical
Subcortical
Subcortical
3.4
4.0
2.4
1.48
1.81
2.02
2.11
1.00
1.67
Subcortical
2.8
1.91
1.90
Subcortical
Cortical and subcortical
Subcortical
Cortical and subcortical
3.6
3.1
3.4
3.7
3.2
2.9
Cortical and subcortical 3.7
Subcortical
3.7
Cortical
3.7
2.8
Subcortical
2.6
2.97
2.05
0.72
1.62
1.41
1.45
2.28
2.95
1.62
0.52
0.72
0.58
1.00
3.08
4.71
1.36
0.76
0.86
—
1.45
2.50
3.33
Cortical and subcortical 3.9
1.47
1.87
3.8
Cortical and subcortical 2.9
3.12
0.64
0.46
0.94
Cortical and subcortical 3.8
Cortical and subcortical 3.6
Cortical and subcortical 2.0
2.81
1.48
0.40
6.45
2.38
2.67
Subcortical
3.6
2.14
1.10
3.32 ⫾ 0.55 1.81 ⫾ 1.04 1.90 ⫾ 1.42
F, Female; M, male; L, left hemisphere; R, right hemisphere; NIHSS, NIH Stroke Scale; SAE, subcortical arteriosclerotic encephalopathy; MCI, middle cerebral artery; PICA, posterior inferior cerebellar artery; WMFT, Wolf Motor Function Test;
MAL, Motor Activity Log; Sensory deficit, paretic hand/control hand.
*Median value.
†
No MRI, lesion site is based on previous diagnostic findings.
Grating orienting task. In accordance with a previous publication
(Weiss et al., 2011), the grating orienting task (GOT) was used to measure limits of tactile resolution, using a modified technique adapted from
Van Boven and Johnson (1994) (see also Craig et al., 2000; Tremblay et
al., 2003; Bleyenheuft and Thonnard, 2007). We used a set of 14 hemispherical plastic domes with gratings cut into their surfaces, resulting in
parallel bars and grooves of equal widths (0.5–10 mm) on each dome.
Before measurement, patients were familiarized with the task by examining and testing the gratings. During sensory testing a cardboard screen
was placed over the patients’ arm to prevent viewing of their hand. Grat-
ings were applied with the ridges and grooves randomly oriented in one
of two orthogonal directions (perpendicular or parallel to the axis of the
finger). Patients were asked to identify the alignment. We determined
GOT thresholds, defined as the groove width at which responses were
75% correct.
Shape-sorter-drum task. The shape-sorter-drum task (SSDT) was used
to measure movement performance (Weiss et al., 2011). Subjects were
required to take 20 objects with their stroke-affected hand from a standard position and put them into a drum that contained several slots into
which objects fitted according to their size and shape. Objects and slots
Sens et al. • Learning and Cortical Reorganization Induced by TFD
11776 • J. Neurosci., August 22, 2012 • 32(34):11773–11779
Table 2. Mean and SD of raw data of the dependent variables VFHT, ED, dipole
moment, SSDT, and GOT for the treatments TFD and placebo
Pre
Post
Variable
Treatment
Mean
SD
Mean
SD
VFHT
TFD
Placebo
TFD
Placebo
TFD
Placebo
TFD
Placebo
TFD
Placebo
66.28
59.39
10.90
10.84
31.27
33.51
5.08
4.87
335.80
329.33
119.49
81.49
4.86
4.18
13.63
19.81
3.24
3.19
341.64
341.82
482.35
74.61
11.95
9.97
37.17
30.67
4.40
4.80
320.26
334.33
245.97
112.81
4.26
4.11
21.92
13.39
3.22
3.08
346.93
342.91
ED
Dipole moment
GOT
SSDT
Figure 2. Changes and SEs in the VFHT thresholds at the forearm during TFD with anesthetic
cream (black) and during placebo (gray). Performance is expressed as a percentage of pretest
value; higher values indicate higher thresholds, i.e., less sensitivity.
were adapted to the abilities of subjects but were the same for a given subject
on all measurements. The dependent variable, performance time, was measured from the start of movements to the instant the last object was successfully put into the drum, so shorter times would represent better
performance. SSDT requires visual-motor as well as somatosensory-motor
coordination to successfully pass the task.
Statistical analysis
After a logarithmic transformation of VFHT values, all mentioned outcome
scales were interval-scaled. Therefore, repeated-measures ANOVAs with the
within-subject factors Time (t1 vs t2) and Treatment ( placebo vs anesthetic cream) were performed to assess differences in standardized ED,
and dipole moment, and also to evaluate the outcome of SSDT, logarithmically transformed VFHT, and GOT as a function of TFD. Post hoc
testing of significant results was performed using paired t tests. We expected interactions between factors Time and Treatment. Linear relationship between the changes ( percentage of pretest values) in ED, dipole
moment, GOT, SSDT, logarithmically transformed VFHT, and the sensory deficit of the index finger were assessed with Spearman correlation.
Conditioned by missing normal distribution of some of these differences, nonparametric correlations were used. All statistical tests were
performed with IBM SPSS Statistic for Windows (19.0) and the significance level was set to p ⬍ 0.05.
arm following application of the anesthetic cream. Of interest is
the Time by Treatment interaction that revealed no differences in
VFHT between t1 and t2 for the placebo cream but significantly
higher VFHT at t2 compared with t1 (t ⫽ ⫺10.30; p ⬍ 0.001; Fig.
2) for the analgesic cream. Furthermore, there were no significant
differences at t1 for the placebo compared with TFD.
Neuromagnetic source imaging
Twenty-nine of the 36 subjects fulfilled the criteria (goodness-offit ⬎ 0.85, dipole moment ⬎ 5 nA m) consistently over all conditions for the SEFs above the contralateral SI. Analysis of
standardized values of ED for D1 and D5 dipoles revealed a significant factor Treatment (F(1,28) ⫽ 4.82; p ⬍ 0.05; ŋ 2 ⫽ 0.15) and
a significant interaction of factors Time and Treatment (F(1,28) ⫽
4.82; p ⬍ 0.05; ŋ 2 ⫽ 0.15). Although the main effect of treatment
resulted from larger ED during TFD, there were no significant
differences of ED at t1 between the two treatments. However, post
hoc analysis of the interaction revealed no significant change between t1 and t2 for the placebo treatment while ED increased
significantly from t1 to t2 when subjects were treated with the
analgesic cream (t ⫽ ⫺2,63; p ⬍ 0.05; Fig. 3A).
In addition, we also analyzed changes in the amount of cortical activity based on standardized values of the dipole moment
that represented activity of D1. ANOVA showed a significant
factor Treatment (F(1,29) ⫽ 7.98; p ⬍ 0.05; ŋ 2 ⫽ 0.22). The Treatment effect resulted from overall significantly stronger dipole
moments during TFD than during placebo. There was a trend
toward a main effect of factor Time (F(1,29) ⫽ 3.14; p ⬍ 0.09; ŋ 2 ⫽
0.10) resulting from a trend toward higher values during t2. Importantly, there was a significant interaction between Time and
Treatment (F(1,29) ⫽ 7.98; p ⬍ 0.05; ŋ 2 ⫽ 0.22). Post hoc testing
demonstrated that the dipole moment significantly increased
from t1 to t2 during TFD (t ⫽ ⫺2.46; p ⬍ 0.05; Fig. 3B), whereas
no change was observed when subjects were treated with placebo.
Moreover, there was no significant difference at t1 between the
two treatments.
Sensory and motor functions
Effectiveness of the TFD on GOT
ANOVA of the GOT data for the index finger revealed a main
effect of factor Time (F(1,35) ⫽ 4.79; p ⬍ 0.05; ŋ 2 ⫽ 0.12) and a
significant interaction between factors Time and Treatment
(F(1,35) ⫽ 11.229; p ⬍ 0.01; ŋ 2 ⫽ 0.24). Regardless of treatment,
better tactile resolution was observed at t2 compared with t1
(main effect of Time). The interaction indicates that the improvement of tactile resolution was stronger during TFD due to
significantly better tactile resolution at t2 than at t1 when subjects
were treated with the analgesic cream (t ⫽ 3.02; p ⬍ 0.01; Fig. 4).
However, no significant differences of tactile resolution were ob-
Results
Mean values and SDs of raw data for the dependent variables
VFHT, ED, Dipole Moment, GOT, and SSDT for the TFD and
placebo treatments are presented in Table 2.
Effectiveness of the TFD on VFHT
We found significant main effects of factors Time (F(1,35) ⫽ 98.19;
p ⬍ 0.001; ŋ 2 ⫽ 0.74) and Treatment (F(1,35) ⫽ 40.25; p ⬍ 0.001;
ŋ 2 ⫽ 0.54) on VFHT scores, as well as a significant Time by
Treatment interaction (F(1,35) ⫽ 84.31; p ⬍ 0.001; ŋ 2 ⫽ 0.71).
The main effect of Time resulted from higher thresholds at t2
compared with t1 in both treatments. The main effect of Treatment indicates overall higher thresholds during TFD on the fore-
Figure 3. Changes and SEs in the ED between D1 and D5 (A) and of the dipole moment for D1
(B) during TFD with anesthetic cream (black) and during placebo (gray). Changes are expressed
as a percentage of pretest value.
Sens et al. • Learning and Cortical Reorganization Induced by TFD
Figure 4. Thresholds and SEs in GOT performance during TFD with anesthetic cream (black)
and with the placebo (gray) for the more-affected index finger. Performance is expressed as a
percentage of pretest value; lower values indicate lower thresholds, i.e., better performance.
The inset shows the task structure.
J. Neurosci., August 22, 2012 • 32(34):11773–11779 • 11777
Figure 6. Scatter plot illustrating the relationship (negative correlation) between changes in
SSDT scores and changes in ED during TFD. Changes are expressed as a percentage of pretest
value.
relation between the anesthetic effect of the cream on the
stroke-effected forearm and the sensory deficit of the index
finger (r ⫽ ⫺0.13; p ⬎ 0.01).
Discussion
Figure 5. Mean performance (time needed and SEs) for the SSDT during TFD with anesthetic
cream (black) and during placebo (gray). Performance is expressed as a percentage of pretest
value; lower values indicate lower thresholds, i.e., better performance. The inset shows the task
structure.
tained at t1 between the two treatments as well as between t1 and
t2 when subjects were treated with the placebo. Furthermore, the
index finger of the stroke-affected hand showed a significant sensory deficit (tactile resolutions, mean ⫾ SD: 4.83 ⫾ 3.09 mm)
compared with the index finger of the other hand (2.82 ⫾ 1.26
mm; t ⫽ 3.94; p ⬍ 0.01). The results remain essentially the same
when analyzing the 29 patients fulfilling the MEG criteria.
Effectiveness of the TFD on SSDT
ANOVA revealed no significant main effects of factors Time or
Treatment (all P values ⬎ 0.1), but a significant interaction
between factors Time and Treatment (F(1,35) ⫽ 4.53; p ⬍ 0.05; ŋ 2
⫽ 0.12). Analysis of the interaction revealed no differences in
performance times at t1 compared with t2 when subjects were
treated with placebo cream, but significantly shorter performance times at t2 when subjects were treated with the analgesic
cream compared with t1 (t ⫽ 2.88; p ⬍ 0.05; Fig. 5). Moreover,
there were no significant differences between placebo and TFD at
t1. The results remain essentially the same when analyzing only
the 29 patients fulfilling the MEG criteria.
Relationship between parameters
We examined the linear relationship between improvements in
cortical activation, somatosensory capacity, and motor performance. All data were standardized with respect to the baseline
value (t1). Changes in ED significantly correlated negatively with
changes in SSDT (r ⫽ ⫺0.46; p ⬍ 0.05; Fig. 6), i.e., a greater ED is
related to better motor performance. There were no significant
correlations between the increase of dipole moment, improvement of GOT, and improvement of SSDT.
We also tested whether the stroke-induced sensory deficits
interact with the anesthetic effect of TFD. There was no cor-
The main findings of this study are threefold: (1) TFD of the
stroke-affected forearm of chronic post-stroke patients led to
cortical plasticity in the contralateral SI, i.e., the Euclidian distance between dipole locations of SEFs to stimulation of D1 and
D5 increased from baseline to post-treatment following TFD but
not following placebo treatment; (2) TFD led to an improvement
of motor performance; and (3) TFD resulted in improved somatosensory sensitivity in the more-affected hand.
Effectiveness of temporary functional deafferentation
In the present study, we used a pharmacologically induced TFD
on the stroke-affected forearm of chronic stroke patients to improve their sensory and motor abilities as well as to modulate
cortical reorganization in the contralateral SI. TFD was highly
efficient in reducing the somatosensory sensitivity of the affected
forearm. Moreover, TFD with an anesthetic cream is an inexpensive technique with only minimal side-effects in the form of
numbness in the affected forearm.
In contrast to other methods of deafferentation, TFD with an
anesthetic cream has some advantages. Tourniquet-induced
deafferentation is a painful procedure and leaves patients without
motor control of the hand. Nerve blocks (e.g., brachial block) are
invasive and often not tolerated by the patients, especially when
applied on a daily basis in the clinical routine. In contrast, TFD
with an anesthetic cream could be used during the whole training
period of CIMT with longer follow-up periods. Rosén et al.
(2006) showed improved sensory relearning in nerve-injured patients 6 weeks after a daily treatment with TFD over a period of 2
weeks. Hand motor practice during a single pharmacologically
induced anesthesia resulted in improved motor function in
stroke patients 2 weeks after treatment (Muellbacher et al., 2002).
However, data on long-term effects of a prolonged application of
TFD by an anesthetic cream are still missing for stroke patients.
Effects of TFD on somatosensory-evoked magnetic fields
TFD resulted in an increased ED of cortical sources of D1 and D5
for the stroke-affected hand in the contralateral SI. Larger distances were found within 5 h following TFD but not following
placebo treatment. This observation might be due to the expansion of the hand area of SI into the representational area of the
stroke-affected forearm whose sensory input was temporarily interrupted by TFD. This is in line with previous reports on healthy
subjects where the neural representation area of the deafferented
Sens et al. • Learning and Cortical Reorganization Induced by TFD
11778 • J. Neurosci., August 22, 2012 • 32(34):11773–11779
body part in SI was invaded by neighboring body parts (Weiss et
al., 2004; Björkman et al., 2009). This result also corresponds to
animal studies demonstrating short- and long-term changes in SI
representation areas of different body parts following deafferentation (Merzenich et al., 1984; Calford and Tweedale, 1988, 1991;
Pons et al., 1991).
TFD also resulted in a significantly increased dipole moment
for the ECD of D1. This result is in line with previous findings in
healthy subjects (Rossini et al., 1994; Björkman et al., 2004b).
These changes might be explained by unmasking of existing silent
connections based on a decrease of intracortical inhibition
between adjacent cortical representations (Rossini et al., 1994;
Liepert et al., 2004; Weiss et al., 2004; Floel et al., 2008), probably mediated through regulation of GABAergic transmission
(Levy et al., 2002; Werhahn et al., 2002a). Thus, TFD might be
used to influence and modulate cortical plasticity in stroke
rehabilitation.
Effects of TFD on sensory function
TFD on the stroke-affected forearm significantly improved somatosensory sensitivity as measured by GOT. This is in accordance with previous results in healthy subjects demonstrating
somatosensory improvements following TFD with anesthetic
cream (Björkman et al., 2004a; Björkman et al., 2009) as well as
with other kinds of TFD (Werhahn et al., 2002b; Weiss et al.,
2004). Furthermore, different types of TFD were also shown to
enhance the somatosensory sensibility of the stroke-affected
hand in stroke patients (Voller et al., 2006; Weiss et al., 2011). As
mentioned above, the observed improvements are probably
based on rapid changes in the receptive fields at different levels of
the somatosensory system, e.g., in the somatosensory cortex as
well as in the thalamus (Nicolelis et al., 1993; Weiss et al., 2004;
Jain et al., 2008; Björkman et al., 2009). Moreover, pharmacologically induced TFD on the right hand of healthy subjects was
found to increase the central processing of the somatosensory
system as indicated by increased somatosensory evoked potentials (Tinazzi et al., 2003). Overall, we suggest that TFD of the
stroke-affected forearm is accompanied by rapid changes in the
receptive fields of the somatosensory system. Our results suggest
that TFD might enhance somatosensory capacities in chronic
post-stroke patients.
Effects of TFD on motor function
In this study we also observed an improved motor performance
as measured by SSDT during TFD compared with placebo treatment. Improving motor function is one of the main aims of neurorehabilitation for stroke patients (Langhorne et al., 2009;
Cramer et al., 2011). A few experimental studies on TFD have
addressed this problem and showed that increased somatosensory sensibility contributes to improvement of motor function in
healthy subjects (Björkman et al., 2004b) as well as in post-stroke
patients (Muellbacher et al., 2002; Floel et al., 2004; Floel et al.,
2008). The results of this study are in line with our previous
finding in stroke patients (Weiss et al., 2011) and demonstrate
that TFD might help to improve motor functions.
Relationship between parameters
We found a significant relationship between cortical reorganization and motor function. Thus, an improved motor performance
is associated with a larger distance of cortical sources of D1 and
D5 in SI. Recent experimental studies demonstrated improved
motor outcomes and changes in interhemispheric plasticity
(Björkman et al., 2004b; Floel et al., 2004). In addition to these
results we found a relationship between improved motor performance of the stroke-affected forearm and changes in the contralateral (ipsilesional) SI. This may be due to the high functional
and anatomical interconnection between the motor and somatosensory system (Pavlides et al., 1993; van Meer et al., 2010).
We also expected an improvement in motor function associated with an increase of somatosensory sensibility. However,
contrary to our hypothesis we did not find changes in GOT related to improvements in SSDT, thus no significant linear relationship between sensory and motor improvement. A reason for
this unexpected result might lie in the variability of patients’ lesions, including cortical, subcortical, and diffuse areas (Table 1).
Further research is required to elucidate the impact of lesion site
on this relationship. Alternatively, the motor and somatosensory
system could independently benefit from TFD. While previous
studies in healthy subjects (Björkman et al., 2004b) and nerveinjured patients (Björkman et al., 2005) have shown effects of
TFD both on the motor and the somatosensory systems, the relationship between systems during TFD has not been investigated
extensively in healthy subjects nor in chronic patients.
Conclusion
Our data show that a simple, inexpensive, pharmacological
method of TFD of a stroke-affected forearm by an anesthetic
cream results in cortical reorganization as well as in improvements of motor performance and somatosensory discrimination.
After testing of its short-term efficiency, long-term effects of repeated TFD procedures across a period of several days are needed
to ascertain whether this technique might become an additional
tool in the improvement of motor rehabilitation in post-stroke
patients.
References
Bauder H, Taub E, Miltner WHR (2001) Behandlung motorischer Störungen nach Schlaganfall. Die Taubsche Bewegungsinduktionstherapie. Göttingen: Hogrefe.
Baumgärtner U, Magerl W, Klein T, Hopf HC, Treede RD (2002) Neurogenic hyperalgesia versus painful hypoalgesia: two distinct mechanisms of
neuropathic pain. Pain 96:141–151.
Björkman A, Rosén B, Lundborg G (2004a) Acute improvement of hand
sensibility after selective ipsilateral cutaneous forearm anaesthesia. Eur
J Neurosci 20:2733–2736.
Björkman A, Rosén B, van Westen D, Larsson EM, Lundborg G (2004b)
Acute improvement of contralateral hand function after deafferentation.
Neuroreport 15:1861–1865.
Björkman A, Rosén B, Lundborg G (2005) Enhanced function in nerveinjured hands after contralateral deafferentation. Neuroreport
16:517–519.
Björkman A, Weibull A, Rosén B, Svensson J, Lundborg G (2009) Rapid
cortical reorganisation and improved sensitivity of the hand following
cutaneous anaesthesia of the forearm. Eur J Neurosci 29:837– 844.
Bleyenheuft Y, Thonnard JL (2007) Tactile spatial resolution measured
manually: a validation study. Somatosens Mot Res 24:111–114.
Calford MB, Tweedale R (1988) Immediate and chronic changes in responses of somatosensory cortex in adult flying-fox after digit amputation. Nature 332:446 – 448.
Calford MB, Tweedale R (1991) Immediate expansion of receptive-fields of
neurons in area-3b of macaque monkeys after digit denervation. Somatosens Mot Res 8:249 –260.
Craig JC, Johnson KO, Johnston KO (2000) The two-point threshold: not a
measure of tactile spatial resolution. Curr Direct Psychol Sci 9:29 –32.
Cramer SC, Sur M, Dobkin BH, O’Brien C, Sanger TD, Trojanowski JQ,
Rumsey JM, Hicks R, Cameron J, Chen D, Chen WG, Cohen LG, deCharms C, Duffy CJ, Eden GF, Fetz EE, Filart R, Freund M, Grant SJ, Haber
S, et al. (2011) Harnessing neuroplasticity for clinical applications.
Brain 134:1591–1609.
Feigin VL, Lawes CMM, Bennett DA, Barker-Collo SL, Parag V (2009)
Sens et al. • Learning and Cortical Reorganization Induced by TFD
Worldwide stroke incidence and early case fatality reported in 56
population-based studies: a systematic review. Lancet Neurol 8:355–369.
Floel A, Nagorsen U, Werhahn KJ, Ravindran S, Birbaumer N, Knecht S,
Cohen LG (2004) Influence of somatosensory input on motor function
in patients with chronic stroke. Ann Neurol 56:206 –212.
Floel A, Hummel F, Duque J, Knecht S, Cohen LG (2008) Influence of somatosensory input on interhemispheric interactions in patients with
chronic stroke. Neurorehabil Neural Repair 22:477– 485.
Grefkes C, Fink GR (2011) Reorganization of cerebral networks after stroke:
new insights from neuroimaging with connectivity approaches. Brain
134:1264 –1276.
Hummel F, Celnik P, Giraux P, Floel A, Wu WH, Gerloff C, Cohen LG
(2005) Effects of non-invasive cortical stimulation on skilled motor
function in chronic stroke. Brain 128:490 – 499.
Jain N, Qi HX, Collins CE, Kaas JH (2008) Large-scale reorganization in the
somatosensory cortex and thalamus after sensory loss in macaque monkeys. J Neurosci 28:11042–11060.
Langhorne P, Coupar F, Pollock A (2009) Motor recovery after stroke: a
systematic review. Lancet Neurol 8:741–754.
Levy LM, Ziemann U, Chen R, Cohen LG (2002) Rapid modulation of
GABA in sensorimotor cortex induced by acute deafferentation. Ann
Neurol 52:755–761.
Liepert J, Weiss T, Meissner W, Steinrücke K, Weiller C (2004) Exerciseinduced changes of motor excitability with and without sensory block.
Brain Res 1003:68 –76.
Merzenich MM, Nelson RJ, Stryker MP, Cynader MS, Schoppmann A, Zook
JM (1984) Somatosensory cortical map changes following digit amputation in adult monkeys. J Comp Neurol 224:591– 605.
Miltner WHR, Bauder H, Sommer M, Dettmers C, Taub E (1999) Effects of
constraint-induced movement therapy on patients with chronic motor
deficits after stroke: a replication. Stroke 30:586 –592.
Muellbacher W, Richards C, Ziemann U, Wittenberg G, Weltz D, Boroojerdi
B, Cohen L, Hallett M (2002) Improving hand function in chronic
stroke. Arch Neurol 59:1278 –1282.
Nicolelis MAL, Lin RCS, Woodward DJ, Chapin JK (1993) Induction of
immediate spatiotemporal changes in thalamic networks by peripheral
block of ascending cutaneous information. Nature 361:533–536.
Pavlides C, Miyashita E, Asanuma H (1993) Projection from the sensory to
the motor cortex is important in learning motor-skills in the monkey.
J Neurophysiol 70:733–741.
Pons TP, Garraghty PE, Ommaya AK, Kaas JH, Taub E, Mishkin M (1991)
Massive cortical reorganization after sensory deafferentation in adult macaques. Science 252:1857–1860.
Preissler S, Dietrich C, Meissner W, Huonker R, Hofmann GO, Miltner WH,
J. Neurosci., August 22, 2012 • 32(34):11773–11779 • 11779
Weiss T (2011) Brachial plexus block in phantom limb pain: a case report. Pain Med 12:1649 –1654.
Rolke R, Magerl W, Campbell KA, Schalber C, Caspari S, Birklein F, Treede
RD (2006) Quantitative sensory testing: a comprehensive protocol for
clinical trials. Eur J Pain 10:77– 88.
Rosén B, Björkman A, Lundborg G (2006) Improved sensory relearning
after nerve repair induced by selective temporary anaesthesia—a new
concept in hand rehabilitation. J Hand Surg Br 31:126 –132.
Rossini PM, Martino G, Narici L, Pasquarelli A, Peresson M, Pizzella V,
Tecchio F, Torrioli G, Romani GL (1994) Short-term brain plasticity in
humans—transient finger representation changes in sensory cortex somatotopy following ischemic anesthesia. Brain Res 642:169 –177.
Tinazzi M, Rosso T, Zanette G, Fiaschi A, Aglioti SM (2003) Rapid modulation of cortical proprioceptive activity induced by transient cutaneous
deafferentation: neurophysiological evidence of short-term plasticity
across different somatosensory modalities in humans. Eur J Neurosci
18:3053–3060.
Tremblay F, Wong K, Sanderson R, Coté L (2003) Tactile spatial acuity in
elderly persons: assessment with grating domes and relationship with
manual dexterity. Somatosens Mot Res 20:127–132.
Van Boven RW, Johnson KO (1994) The limit of tactile spatial resolution in
humans: grating orientation discrimination at the lip, tongue, and finger.
Neurology 44:2361–2366.
van Meer MP, van der Marel K, Wang K, Otte WM, El Bouazati S, Roeling TA,
Viergever MA, Berkelbach van der Sprenkel JW, Dijkhuizen RM (2010)
Recovery of sensorimotor function after experimental stroke correlates
with restoration of resting-state interhemispheric functional connectivity. J Neurosci 30:3964 –3972.
Voller B, Flöel A, Werhahn KJ, Ravindran S, Wu CW, Cohen LG (2006)
Contralateral hand anesthesia transiently improves post-stroke sensory
deficits. Ann Neurol 59:385–388.
Weiss T, Miltner WHR, Liepert J, Meissner W, Taub E (2004) Rapid functional plasticity in the primary somatomotor cortex and perceptual
changes after nerve block. Eur J Neurosci 20:3413–3423.
Weiss T, Sens E, Teschner U, Meissner W, Preul C, Witte OW, Miltner WHR
(2011) Deafferentation of the affected arm a method to improve rehabilitation? Stroke 42:1363–1370.
Werhahn KJ, Mortensen J, Kaelin-Lang A, Boroojerdi B, Cohen LG (2002a)
Cortical excitability changes induced by deafferentation of the contralateral hemisphere. Brain 125:1402–1413.
Werhahn KJ, Mortensen J, Van Boven RW, Zeuner KE, Cohen LG (2002b)
Enhanced tactile spatial acuity and cortical processing during acute hand
deafferentation. Nat Neurosci 5:936 –938.